As veteran of the IT industry, I fall inline with the 10,000 hour rule in reference to Linux/Unix Systems Administration. Needless to say, as a wrangler of *nix based operating systems, most people would consider me an expert; although, I rarely feel that way. However, as of recent I’ve been doing a lot of work on Big Data platforms, including Hadoop, Elastic Search, MongoDB, and using tools such as Storm and Impala. While the install of these platforms is trivial, I can unequivocally say that the tuning is not. This brings me to the point, The Case for Managed Services. I’m fortunate enough to be working with some very sharp guys in the Big Data industry right now, the folks over at InfoChimps. They have developed a Big Data managed service platform that is, in my opinion, unrivaled in the industry. This is the type of platform that it would take months, possibly years, and a team of systems engineers and developers to perfect; and that is the argument I make.
Even though I recently wrote a blog series on how to select a managed service provider, I generally tend to stay away from managed services. All things being equal, I would rather just build the data center and respective application stacks out myself. But, even as a veteran of open source operating systems and software, I would be intimidated to take on the task of building the platform that they have nearly perfected. I don’t say this lightly, and it’s not an attempt to drum up work for myself or my buddies over at InfoChimps, but as a reminder to us all. The reminder I speak of can be summed up in two common mantras that were drilled into me during my time in the Marines. The first: Work smarter, not harder. And the second: Use the right tool for the job.
Given the ridiculous speed of evolving technologies, we can’t all be experts in everything, although I know we all try. Remember, when you come up against a proverbial wall, there are good assets out there you can leverage to architect solutions for your problems; this is where managed services come in. I’m not saying we should outsource our entire technology departments, I’m actually not a fan of outsourcing, but I do believe in rare cases where capital spent to a managed service is a great deal more efficient and fiduciarily responsible than trudging through a neverending platform deployment that is just as likely to fail as it is to succeed. So work smarter by deciding whether or not that next project is something you can take on internally; and once the decision is made, use the right tool for the job by leveraging the proper assets, be it your internal technical staff, a consultant, or a managed service.